Introduction
Manufacturing semiconductor chips, such as processors, graphic or memory units, is one of the most complicated and costly industrial manufacturing processes.
Each one of the advanced processor chips that we carry in our phones undergoes thousands of manufacturing steps at fabrication plants (Fabs), where the recipe in each step must be finely tuned to lead to the desired results. This makes the research, development, and testing cycles of semiconductors extremely costly, time-consuming and labour-intensive.
This PhD project aims to develop models and tools that predict nanoscale defects in semiconductor manufacturing, improving yield, reducing development cycles and making the industry more economical and sustainable. As a PhD student, you will be utilising the exceptional fabrication facilities we have at the University of Southampton, the large supercomputer cluster (IRIDIS) and collaborations with experts in AI. You will also work in collaboration with a team in Silvaco, a global industrial leader in fabrication modelling and work as part of the spinout company, Deep Fabrication Ltd., toward commercialising your models to the semiconductor industry.
Method and Facilities
You will work in one of the world’s most advanced university cleanrooms and a comprehensive suite of characterisation labs. You’ll also use the IRIDIS supercomputing cluster and collaborate with AI experts across the university. Industrial links are central: you will interact with Silvaco (a global leader in fabrication modelling) and contribute to a commercial pathway withing our new spin-out company Deep Fabrication Ltd.
Supervisory Team
The supervisory team, Dr Yasir Noori (ECS), Dr Ben Mills (ORC) and Dr Firman Simanjuntak (ECS), have combined an extensive set of expertise in semiconductor fabrication, artificial intelligence for manufacturing, and characterisation of devices and systems.
After completing your PhD you will have a unique skill set that will put you at the forefront of future semiconductor fabrication, making you well positioned to have a fruitful career in academia or industry.
References:
O. Buchnev, et al. “Deep-Learning-Assisted Focused Ion Beam Nanofabrication” Nano Letters, 22, 2734-2739, 2022.
M. Nandipati, et al. “Bridging Nanomanufacturing and Artificial Intelligence – A Comprehensive Review” Materials, 17, 1621, 2024.
G. Tello, et al. “Deep Structures Machine Learning Model for the Recognition of Mixed-Defect Patterns in Semiconductor Fabrication Processes”, IEEE Transactions on Semiconductor Manufacturing, 31, 2, 2018.
M. Maggipinto, et al. “DeepVM: A Deep Learning-based Approach with Automatic Feature Extraction for 2D Input Data Virtual Metrology” Journal of Process Control, 84, 24-34, 2019.
Yijie Liu, et al. “Towards Smart Scanning Probe Lithography: A Framework Acceleration Nanofabrication Process with in-situ Characterisation via Machine Learning” Microsystems & Nanoengineering, 9, 128, 2023.